Unlocking Flexibility in District Heating Networks by Agent-Based Coupling of Scheduling Optimization and Simulation

Abstract:
This paper describes a new approach of cross-sectoral day-ahead scheduling of generation plants for district heating sup-ply, in which the flexibility potential of the district heating network is utilized. Flexibility in this context is achievable due to the dynamic flow characteristic of the network that enables storing energy directly in the district heating network by decoupling the feed-in from the concurrent heat consumption. Therefore, network topology among other aspects, such as cost or specific CO2 emission of all energy generation options and their coupling to the electricity market, must be considered. For day-ahead scheduling, mixed-integer optimization is generally used without taking the physical and thermal-hydraulic properties of the district heating network into account. In this paper, it is outlined how day-ahead scheduling is coupled with a thermal-hydraulic simulation of the district heating network. The coupling is realized using a multi-agent system, which allows the application of the conventional day-ahead scheduling and simulation methods with only minor modifi-cations. In addition, a forecasting agent and a sensor agent are added to the multi-agent system to enable the exchange of all data required for scheduling between the system's agents. This paper describes how the multi-agent system is set up and how the individual agents interact with each other. It discusses which data should be exchanged between the optimi-zation and simulation agents to enable interaction between the models. In addition, possible approaches on how to adapt the scheduling methodology so that the flexibility potential of the district heating network can be optimally used are explained.